Ground cover information extraction method based on hyperspectral and high-resolution remote sensing imagery

Abstract

This invention discloses a method for extracting ground feature information based on hyperspectral and high-resolution remote sensing imagery, relating to the field of remote sensing technology. The method includes: acquiring hyperspectral and high-resolution remote sensing imagery and preprocessing it; cropping and fusing the preprocessed imagery to obtain a fused image; performing spectral analysis and band filtering on the fused imagery to obtain first image data; performing image enhancement processing on the first image data to obtain multiple grayscale images; combining the multiple grayscale images by band, and performing raster analysis and band filtering to obtain second image data; performing threshold segmentation on the second image data and raster transformation to obtain a first vector; acquiring DEM data within a preset elevation range of the study area and performing raster transformation to obtain a second vector; and cropping the first vector based on the second vector to obtain a distribution vector result. According to the method of this invention, the extraction accuracy and reliability can be effectively improved.

Description

Technical Field

[0001] This invention relates to the field of remote sensing image technology, and in particular to a method for extracting ground feature information based on hyperspectral and high-resolution remote sensing images. Background Technology

[0002] Peach trees are an important economic crop, ranking among the top fruit crops in my country in terms of both planting area and yield. Traditional methods of obtaining spatial distribution information for peach orchards require manual field surveys, consuming significant human, material, and financial resources. Remote sensing technology, with its rapid data updates, wide monitoring range, low cost, and rich information content, is currently an important means of acquiring geographic information. The continuous development of remote sensing technology has made it technically possible to quickly and accurately obtain information on large-scale fruit tree planting. Utilizing remote sensing technology to extract peach orchard area data is of great significance for peach planting planning and yield prediction.

[0003] Hyperspectral remote sensing is a novel remote sensing technology that acquires spectral information of ground features within a continuous spectral range, enabling precise classification and identification of these features. Compared to traditional multispectral remote sensing imagery, hyperspectral imagery offers higher spectral resolution and richer ground feature information, providing more accurate and comprehensive data support for peach tree extraction. High-resolution imagery clearly displays detailed information about ground features, allowing for more accurate extraction of the location and texture characteristics of peach trees. Fusion of hyperspectral and high-resolution imagery improves the accuracy and reliability of peach tree extraction and reduces interference from other ground features in the extraction of peach planting areas.

[0004] Currently, there is limited research on remote sensing extraction of peach trees, while conventional remote sensing image land cover extraction mostly employs classification methods, including supervised and unsupervised classification. Supervised classification methods require a large number of training samples, defining different categories based on the characteristics of the training samples, and then classifying each pixel. This method requires multiple manual interventions, resulting in high time costs. Furthermore, it is crucial to ensure the representativeness, completeness, and quality of the training samples; otherwise, errors may be introduced. Unsupervised classification does not require pre-selection of training samples but classifies based on the similarity between pixels. However, its classification accuracy is relatively low, and it is more sensitive to noise and outliers, easily leading to misclassification. Therefore, a rapid and effective technical solution is urgently needed to address these issues. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a method, electronic device, and storage medium for extracting ground feature information based on hyperspectral and high-resolution remote sensing imagery, which can improve the extraction accuracy, efficiency, and reliability.

[0006] In a first aspect, the method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to embodiments of the present invention includes the following steps:

[0007] Acquire hyperspectral and high-resolution remote sensing images of the study area, and preprocess the hyperspectral and high-resolution remote sensing images.

[0008] The preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image are cropped and fused to obtain a fused image;

[0009] Spectral analysis is performed on the fused image, and band filtering is performed on the fused image based on the analysis results to obtain the first image data of a single band;

[0010] The first image data is subjected to multiple image enhancement processes, and a corresponding grayscale image is obtained after each image enhancement process.

[0011] Multiple grayscale images are combined by band to obtain a multi-band image;

[0012] Raster analysis and band filtering are performed on the multi-band images to obtain single-band second image data;

[0013] The target features are segmented from the second image data into raster results, and the raster results are rasterized to obtain a first vector.

[0014] Obtain DEM data within a preset elevation range of the study area, and perform rasterization on the DEM data to obtain a second vector;

[0015] Based on the second vector, the first vector is cropped to obtain the distribution vector result of the target land features within the study area.

[0016] According to some embodiments of the present invention, the preprocessing of the hyperspectral remote sensing image and the high-resolution remote sensing image includes:

[0017] Perform at least one of the following on the hyperspectral remote sensing image: band combination, radiometric calibration, atmospheric correction, and geometric correction;

[0018] Perform at least one of radiometric calibration, atmospheric correction, image fusion, and geometric correction on the high-resolution remote sensing image.

[0019] According to some embodiments of the present invention, the step of cropping and fusing the preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image to obtain a fused image includes:

[0020] Based on the boundary vector of the study area, the preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image are cropped.

[0021] All bands of the cropped hyperspectral remote sensing image and the preset bands of the high-resolution remote sensing image are fused to obtain a fused image.

[0022] According to some embodiments of the present invention, the step of performing spectral analysis on the fused image and performing band filtering on the fused image based on the analysis results to obtain single-band first image data includes:

[0023] Acquire the spectral information of target features and non-target features in the fused image, and plot spectral curves for each.

[0024] Based on the spectral curve, the single band with the largest difference in spectral reflectance between the target feature and the non-target feature is selected as the first image data.

[0025] According to some embodiments of the present invention, the step of performing multiple image enhancement processes on the first image data, and obtaining a corresponding grayscale image after each image enhancement process, includes:

[0026] At least two of the following are used to perform multiple image enhancement processes on the first image data: convolution, mean filtering, focus analysis, dilation, erosion, opening operation, and closing operation. A corresponding grayscale image is obtained after each image enhancement process.

[0027] According to some embodiments of the present invention, the step of performing raster analysis and band filtering on the multi-band image to obtain single-band second image data includes:

[0028] Analyze the pixel value distribution patterns of target and non-target features on the multi-band imagery;

[0029] The grayscale image with the highest separation between the target feature and the non-target feature is selected according to preset rules and used as the second image data.

[0030] According to some embodiments of the present invention, the preset rules include:

[0031] The pixel value distribution ranges of the target feature and the non-target feature do not intersect;

[0032] The minimum pixel value of the target feature is less than the minimum pixel value of the non-target feature, or the maximum pixel value of the target feature is greater than the maximum pixel value of the non-target feature.

[0033] According to some embodiments of the present invention, the method further includes:

[0034] The distribution vector results are then subjected to patch deletion and smoothing processing;

[0035] The distribution vector result after smoothing is corrected.

[0036] In a second aspect, an electronic device according to embodiments of the present invention includes:

[0037] Memory, used to store program instructions;

[0038] The processor is configured to call program instructions stored in the memory and execute the method for extracting ground feature information based on hyperspectral and high-resolution remote sensing images as described in the first aspect embodiment according to the obtained program instructions.

[0039] Thirdly, according to an embodiment of the present invention, a computer-readable storage medium stores computer-executable instructions for causing a computer to perform the method for extracting ground feature information based on hyperspectral and high-resolution remote sensing images as described in the first aspect embodiment.

[0040] The method, electronic device, and storage medium for extracting ground feature information based on hyperspectral and high-resolution remote sensing imagery according to embodiments of the present invention have at least the following beneficial effects: They enable the extraction of large-scale peach orchard areas using both hyperspectral and high-resolution remote sensing imagery, providing a wide monitoring range and reducing the workload of manual field investigations. An automated extraction process is established, improving monitoring efficiency. Hyperspectral data is rich in spectral information, while high-resolution imagery provides clear texture and structural information of ground features. The fused image not only has high spatial resolution but also rich spectral information. Extracting peach orchard areas based on the fused imagery effectively improves the accuracy and reliability of peach orchard extraction, reduces interference from other ground features, and provides more accurate and efficient support for applications such as peach orchard management and peach yield prediction.

[0041] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0042] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0043] Figure 1 This is a flowchart illustrating the steps of the method for extracting ground cover information based on hyperspectral and high-resolution remote sensing images according to an embodiment of the present invention.

[0044] Figure 2 This is a flowchart illustrating the method for extracting ground cover information based on hyperspectral and high-resolution remote sensing images according to an embodiment of the present invention.

[0045] Figure 3 This is a false-color display image of the fused image according to an embodiment of the present invention;

[0046] Figure 4 This is a pixel value distribution diagram of peach trees and non-peach trees according to an embodiment of the present invention;

[0047] Figure 5 This is a diagram showing the effect of erosion image processing according to an embodiment of the present invention;

[0048] Figure 6 This is a preliminary extracted vector image of a peach tree according to an embodiment of the present invention;

[0049] Figure 7 This is a DEM image of the study area in this embodiment of the invention;

[0050] Figure 8 This is the final vector image of the peach tree in an embodiment of the present invention. Detailed Implementation

[0051] The embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0052] In the description of this invention, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0053] The terms "first," "second," "third," and "fourth," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0054] In this invention, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0055] On the one hand, embodiments of the present invention propose a method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery, such as... Figure 1 and Figure 2 As shown, the method includes:

[0056] Step S100: Acquire hyperspectral and high-resolution remote sensing images of the study area, and preprocess the hyperspectral and high-resolution remote sensing images.

[0057] Step S200: Crop and fuse the preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image to obtain a fused image;

[0058] Step S300: Perform spectral analysis on the fused image and filter the fused image by band based on the analysis results to obtain the first image data of a single band;

[0059] Step S400: Perform multiple image enhancement processes on the first image data, and obtain a corresponding grayscale image after each image enhancement process;

[0060] Step S500: Combine multiple grayscale images into a band composite to obtain a multi-band image;

[0061] Step S600: Perform raster analysis and band selection on the multi-band image to obtain the second image data of a single band;

[0062] Step S700: Segment the target features from the second image data into raster results, and perform rasterization on the raster results to obtain the first vector;

[0063] Step S800: Obtain DEM data within the preset elevation range of the study area, and rasterize the DEM data to obtain the second vector;

[0064] Step S900: Based on the second vector, the first vector is cropped to obtain the distribution vector results of target features within the study area.

[0065] Specifically, hyperspectral remote sensing is a novel remote sensing technology that acquires spectral information of ground features within a continuous spectral range, enabling precise classification and identification of these features. Compared to traditional multispectral remote sensing imagery, hyperspectral imagery offers higher spectral resolution and richer ground feature information, providing more accurate and comprehensive data support for the extraction of peach trees (or other fruit trees / planting areas). High-resolution imagery clearly displays detailed information about ground features, allowing for more accurate extraction of the location and texture characteristics of peach trees. Fusion of hyperspectral and high-resolution imagery improves the accuracy and reliability of peach tree extraction and reduces interference from other ground features in the extraction of peach planting areas.

[0066] First, hyperspectral and high-resolution remote sensing images of the study area were acquired and preprocessed. Preprocessing included band combination, radiometric calibration, atmospheric correction, and geometric correction for the hyperspectral images, and radiometric calibration, atmospheric correction, image fusion, and geometric correction for the high-resolution images. This preprocessing facilitates subsequent image processing and analysis.

[0067] Then, based on the boundary vector of the study area, the preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image are cropped. All bands of the cropped hyperspectral remote sensing image and a preset band of the high-resolution remote sensing image are then fused to obtain a fused image. In this example, the preset band of the high-resolution remote sensing image is the near-infrared band, meaning that all bands of the cropped hyperspectral remote sensing image and the near-infrared band of the high-resolution remote sensing image are fused to obtain the fused image.

[0068] After obtaining the fused image, the spectral information of the target and non-target features in the fused image is acquired, and spectral curves are plotted separately. Based on the spectral curves, the single band with the largest difference in spectral reflectance between the target and non-target features is selected as the first image data. Assuming the target feature refers to a peach tree planting area, the spectral information of peach trees and non-peach trees in the fused image is collected, spectral curves are plotted, and then, based on the pattern of the spectral curves of peach trees and non-peach trees, the single band with the largest difference in spectral reflectance between peach trees and non-peach trees is selected as the basic image data for extracting peach trees.

[0069] After obtaining the first image data, multiple image enhancement processes can be performed on it, each yielding a corresponding grayscale image. For example, image enhancement methods such as convolution, mean filtering, focus analysis, dilation, erosion, opening, and closing can be used to process the first image data, obtaining multiple grayscale images. These multiple grayscale images are then combined into a multi-band image; further, raster analysis and band filtering are performed on the multi-band image to obtain single-band second image data. Specifically, raster analysis and band filtering of the multi-band image involve analyzing the pixel value distribution patterns of target and non-target features on the multi-band image; and selecting the grayscale image with the highest separation between target and non-target features according to preset rules as the second image data.

[0070] Assuming the target feature is a peach orchard, the pixel value distribution patterns of peach trees and non-peach tree features are analyzed based on multi-band imagery. Rules are then set to filter out the grayscale images with the highest separation between peach trees and non-peach tree features, which will be used as image data for further peach tree extraction. Preset rules can be set as follows:

[0071] (1) The pixel value distribution ranges of the target feature (peach tree) and non-target features (non-peach tree) do not intersect;

[0072] (2) The minimum pixel value of the target feature (peach tree) is less than the minimum pixel value of the non-target feature (non-peach tree), or the maximum pixel value of the target feature is greater than the maximum pixel value of the non-target feature.

[0073] After obtaining the second image data in a single band, the raster results of the target ground objects are segmented from the second image data, and the raster results are rasterized to obtain the first vector. In this example, a threshold segmentation method can be used to segment the second image data. Specifically, a threshold range is set based on the second image data. The threshold range is determined according to the maximum and minimum pixel values ​​of the target ground objects in the second image data. If the minimum pixel value of the target ground object is 'a' and the maximum pixel value is 'b', then the threshold range is set to [a, b]. All pixels in the second image data that are within the threshold range are extracted, and the raster results of the peach tree planting area are segmented. The raster results of the peach tree planting area are rasterized to obtain the preliminary extracted vector of the peach tree, i.e., the first vector.

[0074] Obtain DEM (Digital Elevation Model) data within a predetermined elevation range of the study area, and perform rasterization on the DEM data to obtain the second vector. Assuming the target feature is a peach orchard area, and since peach trees are mostly planted in plains, the area containing the peach orchard area is defined as the study area. The study area is typically also a plain. Spatial analysis is performed on the DEM data of the study area to extract pixels within a specified elevation range, thus obtaining the DEM data within that specified elevation range. Then, the obtained DEM data within the specified elevation range is rasterized to obtain the plain area vector, i.e., the second vector.

[0075] Based on the second vector, the first vector is clipped to obtain the distribution vector results of target features within the study area. Specifically, using the second vector as the clipping element and the first vector as the input element, the distribution vector results of peach tree planting areas within the plain area are obtained.

[0076] Furthermore, in some embodiments of the present invention, the method for extracting ground cover information based on hyperspectral and high-resolution remote sensing images further includes the following steps:

[0077] Perform patch deletion and smoothing on the distribution vector results;

[0078] The distribution vector results after smoothing are corrected.

[0079] Specifically, the distribution vector results undergo fine patch deletion and smoothing. Fine patch deletion is achieved by calculating the patch area and removing areas smaller than a set range. Smoothing involves smoothing sharp corners in the outline of the distribution vector results to improve vector quality. Finally, the distribution vector results are manually corrected, removing incorrectly extracted patches and outlining missing patches to obtain the distribution vector of the peach tree planting area.

[0080] It should be noted that the method for extracting ground feature information based on hyperspectral and high-resolution remote sensing images according to embodiments of the present invention is applicable not only to the extraction of peach tree planting areas, but also to the extraction of other fruit trees / plants / other ground features.

[0081] The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to embodiments of the present invention utilizes both hyperspectral and high-resolution remote sensing imagery to extract large-scale peach orchard areas. This method provides a wide monitoring range, reduces the workload of manual field investigations, establishes an automated extraction process, and improves monitoring efficiency. Hyperspectral data offers rich spectral information, while high-resolution imagery provides clear texture and structural information of ground features. The fused image not only boasts high spatial resolution but also possesses abundant spectral information. Extracting peach orchard areas based on this fused imagery effectively improves the accuracy and reliability of peach orchard extraction, reduces interference from other ground features, and provides more accurate and efficient support for applications such as peach orchard management and peach yield prediction.

[0082] The following specific embodiment illustrates the method for extracting ground cover information based on hyperspectral and high-resolution remote sensing images according to the present invention. It should be noted that the following embodiment is merely an illustrative example and not a specific limitation of the present invention.

[0083] Step 1: The hyperspectral remote sensing image uses 10-meter resolution hyperspectral satellite data (hereinafter referred to as OHS), and the high-resolution remote sensing image uses high-resolution satellite data with a resolution better than 1 meter (hereinafter referred to as GF-2). The OHS and GF-2 data are preprocessed respectively.

[0084] Step 2: After determining the study area, the preprocessed OHS and GF-2 data are cropped using the boundary vector of the study area, and then the cropped data are fused. In this example, the Gram-Schmidt Pan Sharpening (GS) fusion algorithm is used. Statistical analysis is used to achieve optimal matching of each band involved in the fusion, avoiding the problem of excessive concentration of information in certain bands. This maintains the consistency of spectral information before and after fusion, making it a high-fidelity remote sensing image fusion method. The preprocessed OHS data has 32 bands, and the GF-2 image has four bands: red, green, blue, and near-infrared. The near-infrared band of GF-2 is selected and fused with the 32 bands of OHS using the GS fusion algorithm. The number of bands in the fused image is the same as the number of OHS bands (32 bands), and the resolution is the same as the GF-2 resolution (better than 1 meter). The fused image possesses both the spectral information of the OHS data and the spatial resolution of the GF-2 image, and its false-color display effect is as follows: Figure 3 As shown.

[0085] Step 3: Based on the fused image, collect spectral information of peach trees and non-peach trees, draw spectral curves, and by analyzing the spectral curve patterns of peach trees and non-peach trees, select the 26th band, which has the largest difference in spectral reflectance between peach trees and non-peach trees, as the basic image data for extracting peach trees.

[0086] Step 4: Image enhancement transforms the image into a form more suitable for human eyes or machines to interpret and analyze. For band 26, image enhancement methods such as convolution, mean filtering, focus analysis, dilation, erosion, opening, and closing are applied.

[0087] Convolution divides the entire pixel into blocks and averages them to change the spatial frequency of the image, thereby enhancing it. It can be viewed as a weighted summation process, where each pixel in the image region is multiplied by each element of the convolution kernel, and the sum of all products becomes the new value of the center pixel of the region. Mean filtering uses a neighborhood averaging method, replacing the gray level of each individual pixel with the average of several pixel gray levels. Focus analysis takes the sum, maximum, minimum, mean, median, or standard deviation of all pixels in the neighborhood as the value of the new pixel to achieve image enhancement.

[0088] Dilation is the process of incorporating all background points in contact with an object into that object, causing the boundary to expand outward. It is used to fill in certain voids in the target region and eliminate small particle noise contained within the target region. The formula is as follows:

[0089]

[0090] Erosion is a process that eliminates boundary points, causing the boundary to shrink inward. It can be used to eliminate small and meaningless targets. The formula is as follows:

[0091]

[0092] The closing operation involves first dilating and then eroding the image for filling purposes. The formula is as follows:

[0093]

[0094] The opening operation involves performing erosion followed by dilation to remove isolated small points and eliminate noise. The formula is as follows:

[0095]

[0096] In the formula, A is the input image, B is the structuring element, and (x, y) are the coordinates of the pixel.

[0097] Step 5: After a series of image enhancement processes, band 26 yields 10 grayscale images. These 10 grayscale images are then combined with bands 3, 7, 14, and 26 (or other visible and near-infrared bands) of the fused image. Pixel values ​​of peach tree and non-peach tree ground feature samples are collected, and pixel value distribution curves are plotted (e.g., ...). Figure 4According to the band selection rules, the image after erosion processing was selected as the image for peach tree extraction. Figure 5 ).

[0098] The filtering rules are set as follows:

[0099] (1) The distribution range of peach tree pixel values ​​in the image does not intersect with the distribution range of non-peach tree pixel values;

[0100] (2) The minimum (maximum) value of the peach tree pixel in the image is greater than (less than) the maximum (minimum) value of the non-peach tree pixel.

[0101] Step 6: Define the image pixel values ​​selected in Step 5 as peach trees if they are between 2100 and 3500. Remove pixels outside this threshold to obtain the raster result of the peach tree planting area. Use the "Raster to Vector" tool to convert the raster result of the peach tree planting area into a vector, obtaining the preliminary extracted vector of the peach tree planting area. Figure 6 ).

[0102] Step 7: The DEM data for the study area uses the ASTER GDEM digital elevation model. The data is downloaded according to the boundaries of the study area and then resampled. Figure 7 The "Extract by Attribute" tool was used to extract pixels with elevation values ​​less than or equal to 130, resulting in DEM data for the study area with elevation values ​​not exceeding 130.

[0103] Step 8: Use the "Raster to Vector" tool to convert the DEM data with elevation values ​​not exceeding 130 obtained in Step 7 into vector data to obtain the plain area vector of the study area.

[0104] Step 9: Using the plain area vector obtained in Step 8 as the clipping element and the preliminary extracted vector obtained in Step 6 as the input element, the distribution vector of peach trees within the plain area is clipped to obtain the result.

[0105] Step 10: Perform fine patch deletion and smoothing on the vector results within the plain area. Deleting patches with an area less than 100 square meters achieves fine patch deletion; the PAEK algorithm is used to smooth the vectors with a smoothing tolerance of 5 meters.

[0106] Step 11: Manually modify the smoothed vector, delete misextracted patches, and outline missing patches to obtain the final peach tree planting area distribution vector. Figure 8 ).

[0107] Steps 6-11 can be performed in ArcGIS software.

[0108] On the other hand, embodiments of the present invention also propose an electronic device, comprising:

[0109] Memory, used to store program instructions;

[0110] The processor is used to call program instructions stored in the memory and execute the method for extracting ground cover information based on hyperspectral and high-resolution remote sensing images as described in the above-mentioned embodiments according to the obtained program instructions.

[0111] On the other hand, embodiments of the present invention also provide a storage medium, which is a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the above-described method for extracting ground feature information based on hyperspectral and high-resolution remote sensing images.

[0112] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate, and may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0113] Although specific embodiments are described herein, those skilled in the art will recognize that many other modifications or alternative embodiments are also within the scope of this disclosure. For example, any of the functions and / or processing capabilities described in connection with a particular device or component can be performed by any other device or component. Furthermore, while various exemplary embodiments and architectures have been described according to embodiments of this disclosure, those skilled in the art will recognize that many other modifications to the exemplary embodiments and architectures described herein are also within the scope of this disclosure.

[0114] The foregoing description, with reference to block diagrams and flowcharts of systems, methods, systems, and / or computer program products according to exemplary embodiments, has described certain aspects of this disclosure. It should be understood that one or more blocks in the block diagrams and flowcharts, as well as combinations of blocks in the block diagrams and flowcharts, can be implemented by executing computer-executable program instructions, respectively. Similarly, according to some embodiments, some blocks in the block diagrams and flowcharts may not need to be executed in the order shown, or may not all need to be executed. Furthermore, additional components and / or operations beyond those shown in the blocks in the block diagrams and flowcharts may exist in some embodiments.

[0115] Therefore, blocks in block diagrams and flowcharts support combinations of means for performing a specified function, combinations of elements or steps for performing a specified function, and program instruction means for performing a specified function. It should also be understood that each block in a block diagram and flowchart, and combinations of blocks in block diagrams and flowcharts, can be implemented by a dedicated hardware computer system or a combination of dedicated hardware and computer instructions that performs a specific function, element, or step.

[0116] The program modules, applications, etc., described herein may include one or more software components, including, for example, software objects, methods, data structures, etc. Each such software component may include computer-executable instructions that, in response to execution, cause at least a portion of the functionality described herein (e.g., one or more operations of the exemplary methods described herein) to be performed.

[0117] Software components can be coded using any of a variety of programming languages. An exemplary programming language could be a low-level programming language, such as assembly language associated with a specific hardware architecture and / or operating system platform. Software components including assembly language instructions may need to be converted into executable machine code by an assembler before being executed by the hardware architecture and / or platform. Another exemplary programming language could be a higher-level programming language that is portable across multiple architectures. Software components including higher-level programming languages ​​may need to be converted into an intermediate representation by an interpreter or compiler before execution. Other examples of programming languages ​​include, but are not limited to, macro languages, shell or command languages, job control languages, scripting languages, database query or search languages, or report writing languages. In one or more exemplary embodiments, a software component containing instructions from one of the above-described programming language examples can be executed directly by the operating system or other software components without first being converted into another form.

[0118] Software components can be stored as files or other data storage structures. Software components of similar type or related function can be stored together in a specific directory, folder, or library. Software components can be static (e.g., pre-defined or fixed) or dynamic (e.g., created or modified at runtime).

[0119] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery, characterized in that, Includes the following steps: Acquire hyperspectral and high-resolution remote sensing images of the study area, and preprocess the hyperspectral and high-resolution remote sensing images. The preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image are cropped and fused to obtain a fused image; Spectral analysis is performed on the fused image, and band filtering is performed on the fused image based on the analysis results to obtain the first image data of a single band; The first image data is subjected to multiple image enhancement processes, and a corresponding grayscale image is obtained after each image enhancement process. Multiple grayscale images are combined by band to obtain a multi-band image; Raster analysis and band filtering are performed on the multi-band images to obtain single-band second image data; The target features are segmented from the second image data, and the rasterized results are transformed into polygons to obtain the first vector. Obtain DEM data within a preset elevation range of the study area, and perform rasterization on the DEM data to obtain a second vector; Based on the second vector, the first vector is cropped to obtain the distribution vector result of the target land features within the study area.

2. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 1, characterized in that, The preprocessing of the hyperspectral remote sensing image and the high-resolution remote sensing image includes: Perform at least one of the following on the hyperspectral remote sensing image: band combination, radiometric calibration, atmospheric correction, and geometric correction; Perform at least one of radiometric calibration, atmospheric correction, image fusion, and geometric correction on the high-resolution remote sensing image.

3. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 1, characterized in that, The step of cropping and fusing the preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image to obtain a fused image includes: Based on the boundary vector of the study area, the preprocessed hyperspectral remote sensing image and the high-resolution remote sensing image are cropped. All bands of the cropped hyperspectral remote sensing image and the preset bands of the high-resolution remote sensing image are fused to obtain a fused image.

4. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 1, characterized in that, The step of performing spectral analysis on the fused image and filtering the fused image by band based on the analysis results to obtain single-band first image data includes: Acquire the spectral information of target features and non-target features in the fused image, and plot spectral curves for each. Based on the spectral curve, the single band with the largest difference in spectral reflectance between the target feature and the non-target feature is selected as the first image data.

5. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 1, characterized in that, The step of performing multiple image enhancement processes on the first image data, obtaining a corresponding grayscale image after each image enhancement process, includes: At least two of the following are used to perform multiple image enhancement processes on the first image data: convolution, mean filtering, focus analysis, dilation, erosion, opening operation, and closing operation. A corresponding grayscale image is obtained after each image enhancement process.

6. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 1, characterized in that, The step of performing raster analysis and band filtering on the multi-band image to obtain single-band second image data includes: Analyze the pixel value distribution patterns of target and non-target features on the multi-band imagery; The grayscale image with the highest separation between the target feature and the non-target feature is selected according to preset rules and used as the second image data.

7. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 6, characterized in that, The preset rules include: The pixel value distribution ranges of the target feature and the non-target feature do not intersect; The minimum pixel value of the target feature is less than the minimum pixel value of the non-target feature, or the maximum pixel value of the target feature is greater than the maximum pixel value of the non-target feature.

8. The method for extracting ground cover information based on hyperspectral and high-resolution remote sensing imagery according to claim 1, characterized in that, The method further includes: The distribution vector results are then subjected to patch deletion and smoothing processing; The distribution vector result after smoothing is corrected.

9. An electronic device, characterized in that, include: Memory, used to store program instructions; The processor is configured to call program instructions stored in the memory and execute the method for extracting ground feature information based on hyperspectral and high-resolution remote sensing images according to any one of the obtained program instructions.

10. A storage medium, characterized in that, The storage medium stores computer-executable instructions, which are used to cause a computer to perform the method for extracting ground feature information based on hyperspectral and high-resolution remote sensing imagery as described in any one of claims 1-8.

Images